Paper: arXiv 2501.02963

Authors: Paul Ghelasi, Florian Ziel

Abstract

Electricity price forecasting approaches generally fall into two categories: data-driven models, which learn from historical patterns, or fundamental models, which simulate market mechanisms. We propose a novel and highly efficient data-driven merit order model that integrates both paradigms. The model embeds the classical expert-based merit order as a nested special case, allowing all key parameters, such as plant efficiencies, bidding behavior, and available capacities, to be estimated directly from historical data, rather than assumed. We further enhance the model with critical embedded extensions such as hydro power, cross-border flows and corrections for underreported capacities, which considerably improve forecasting accuracy. Applied to the German day-ahead market, our model outperforms both classic fundamental and state-of-the-art machine learning models. It retains the interpretability of fundamental models, offering insights into marginal technologies, fuel switches, and dispatch patterns, elements which are typically inaccessible to black-box machine learning approaches. This transparency and high computational efficiency make it a promising new direction for electricity price modeling.

Complexity vs Empirical Score

  • Math Complexity: 5.5/10
  • Empirical Rigor: 8.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper presents a highly innovative data-driven merit order framework that unifies fundamental electricity market modeling with statistical parameter estimation. The empirical evaluation on the German day-ahead market is exceptionally thorough, outperforming both traditional fundamental models and advanced machine learning baselines while preserving physical interpretability.

Research Flowchart

  flowchart TD
    A[Research Goal: Hybrid Electricity Price Forecasting Model] --> B[Input Data: Historical Market Data, Fuel Prices, Hydro & Capacity]
    B --> C[Methodology: Data-Driven Merit Order Model]
    C --> D[Model Extensions: Hydro Power, Cross-Border Flows & Capacity Corrections]
    D --> E[Parameter Estimation: Data-Driven Efficiency, Bidding & Capacity Optimization]
    E --> F[Validation: German Day-Ahead Market Application]
    F --> G[Outcomes: Superior Accuracy, High Computational Efficiency & Retained Interpretability]

Cited References (82)

  1. microbenchmark: Accurate Timing Functions (2024) — link
  2. Rcpp: Seamless R and C++ Integration (2025) — link
  3. Data-driven estimation in equilibrium using inverse optimization (2015)
  4. A data-driven bidding model for a cluster of price-responsive consumers of electricity (2016)
  5. Inverse Optimization and Forecasting Techniques Applied to Decision-making in Electricity Markets (2017)
  6. Integrated forecasting and scheduling of implicit demand response in balancing markets using inverse optimization (2023)
  7. Deep inverse reinforcement learning for objective function identification in bidding models (2021)
  8. Cooperative learning for P2P energy trading via inverse optimization and interval analysis (2020) — link
and 74 more
  1. Learning from past bids to participate strategically in day-ahead electricity markets (2019)
  2. Strategic equilibrium bidding for electricity suppliers in a day-ahead market using inverse optimization (2017)
  3. Inverse optimization with kernel regression: Application to the power forecasting and bidding of a fleet of electric vehicles (2021)
  4. Enhancing Power System and Market Operations through Stochastic Modeling, Inverse Optimization, and Machine Learning (2025)
  5. Estimating the Unobservable Components of Electricity Demand Response with Inverse Optimization (2024) — link
  6. Inverse equilibrium analysis of oligopolistic electricity markets (2020)
  7. Short-term forecasting of price-responsive loads using inverse optimization (2017)
  8. Revealing rival marginal offer prices via inverse optimization (2013)
  9. Data-Driven Inverse Optimization with Applications in Electricity Markets (2023)
  10. Data-driven inverse optimization for marginal offer price recovery in electricity markets (2023)
  11. Inverse optimization approach to the identification of electricity consumer models (2021)
  12. Inverse optimization for the recovery of market structure from market outcomes: An application to the MISO electricity market (2017)
  13. Modeling electricity prices: jump diffusion and regime switching (2004)
  14. A building energy models calibration methodology based on inverse modelling approach (2022)
  15. The inverse simulation approach: a focused review of methods and applications (2000)
  16. Model prediction and inverse simulation (2018)
  17. An introduction to the mathematical theory of inverse problems (2011)
  18. Inverse Probleme (2015)
  19. Inverse problems in the mathematical sciences (1993)
  20. Computational intelligence applied to inverse problems in radiative transfer (2023)
  21. Inverse problem theory and methods for model parameter estimation (2005)
  22. A Multi-Objective Inverse (MOIn) Energy System Modelling Method for Guiding Early Technology Development (2025)
  23. From day-ahead to mid and long-term horizons with econometric electricity price forecasting models (2025)
  24. Analyzing and forecasting electricity price using regime-switching models: The case of New Zealand market (2023)
  25. The forecasting performance of a finite mixture regime-switching model for daily electricity prices (2014)
  26. Intra-day and regime-switching dynamics in electricity price formation (2008)
  27. Predicting price spikes in electricity markets using a regime-switching model with time-varying parameters (2006)
  28. A regime switching long memory model for electricity prices (2006)
  29. ASSUME: An agent-based simulation framework for exploring electricity market dynamics with reinforcement learning (2025)
  30. AMIRIS: Agent-based Market model for the Investigation of Renewable and Integrated energy Systems (2023)
  31. OSeMOSYS: the open source energy modeling system: an introduction to its ethos, structure and development (2011)
  32. Calliope: a multi-scale energy systems modelling framework (2018)
  33. Auswirkungen einer Beendigung der Kohleverstromung bis 2038 auf den Strommarkt, CO2-Emissionen und ausgew "a hlte Industrien (2019)
  34. Die deutsche Braunkohlenwirtschaft-Historische Entwicklungen, Ressourcen, Technik, wirtschaftliche Strukturen und Umweltauswirkungen (2017)
  35. From day-ahead to mid and long-term horizons with econometric electricity price forecasting models (2024) — link
  36. mlrMBO : A Modular Framework for Model - Based Optimization of Expensive Black - Box Functions (2017) — link
  37. An optimal mix of conventional power systems in the presence of renewable energy: A new design for the German electricity market (2018)
  38. A novel hybrid deep neural network model for short-term electricity price forecasting (2021)
  39. A hybrid approach for probabilistic forecasting of electricity price (2013)
  40. Price forecasting of day-ahead electricity markets using a hybrid forecast method (2011)
  41. A hybrid forecasting system based on a dual decomposition strategy and multi-objective optimization for electricity price forecasting (2019)
  42. Day-ahead electricity price forecasting based on hybrid regression model (2022)
  43. An adaptive hybrid model for short term electricity price forecasting (2020)
  44. PyPSA: Python for power system analysis (2017) — link
  45. Econometric modelling and forecasting of intraday electricity prices (2020)
  46. Understanding intraday electricity markets: Variable selection and very short-term price forecasting using LASSO (2019)
  47. Forecasting electricity spot-prices using linear univariate time-series models (2004)
  48. Optimal learning with Bernstein online aggregation (2017)
  49. Smoothed bernstein online aggregation for short-term load forecasting in ieee dataport competition on day-ahead electricity demand forecasting: Post-covid paradigm (2022)
  50. Kalman recursions aggregated online (2024)
  51. A stochastic approximation method (1951)
  52. Online learning algorithms (2021)
  53. Forecasting: theory and practice (2022)
  54. The elements of statistical learning: data mining, inference, and prediction (2009)
  55. New introduction to multiple time series analysis (2005)
  56. Introduction to time series and forecasting (2002)
  57. Electricity price forecasting: A review of the state-of-the-art with a look into the future (2014)
  58. Electricity price forecasting using sale and purchase curves: The X-Model (2016)
  59. A review of modelling tools for energy and electricity systems with large shares of variable renewables (2018)
  60. Day-ahead electricity price forecasting with high-dimensional structures: Univariate vs. multivariate modeling frameworks (2018)
  61. Data-driven modeling for long-term electricity price forecasting (2022)
  62. Multi-day-ahead electricity price forecasting: A comparison of fundamental, econometric and hybrid models (2021)
  63. Electricity price forecasting in the short term hybridising fundamental and econometric modelling (2019)
  64. Medium-term probabilistic forecasting of electricity prices: A hybrid approach (2016)
  65. Forecasting power prices using a hybrid fundamental-econometric model (2011)
  66. Modeling the merit order curve of the European energy exchange power market in Germany (2013)
  67. Examining the long term relationships between energy commodities prices and carbon prices on electricity prices using Markov Switching Regression (2022)
  68. The impact of renewable energies on EEX day-ahead electricity prices (2014)
  69. Market-wide impact of renewables on electricity prices in Australia (2022)
  70. The impact of renewables on electricity prices in Germany–An estimation based on historic spot prices in the years 2011–2013 (2016)
  71. The impact of renewables on electricity prices in Germany-An update for the years 2014–2018 (2020)
  72. Modelling German electricity wholesale spot prices with a parsimonious fundamental model–Validation & application (2019)
  73. Are fundamentals enough? Explaining price variations in the German day-ahead and intraday power market (2016)
  74. The plunge in German electricity futures prices–Analysis using a parsimonious fundamental model (2016)

0 of 82 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.